Main objectives of this lecture GREEN IPTV A Resource and Energy Efficient Network for IPTV FERNANDO M. V. RAMOS General Illustrate the usual route taken to make computer science research Specific Exemplify with a resource and energy efficient network for IPTV 1. Look for an opportunity
2. Identify a problem The ability to ask the right question is more than half the battle of finding the answer Thomas J. Watson 3. Analyse state of the art 4. Propose solution 5. Evaluate solution 6. Assess relevance
What is IPTV? A typical IPTV network More details Internet IP Network TV Head End Customer Premises TV All TV channels transmitted always everywhere IP multicast used PIM-SM Static multicast trees No traffic engineering... STB DSLAM PC Why IPTV? New service on top of an IP network Still an infant: around 5 years old The sum of all forms of video (IPTV, VoD, P2P) will account for over 91% of global consumer traffic by 2013 [Cisco] In the US there are already more than 5 million IPTV subscribers, and this number is expected to increase to 15.5 million by 2013 [Piper]
Resource inefficiencies 90% of all TV viewing is restricted to a small selection of channels [Cha et al.][qiu et al.] Number of channels 160 140 120 100 80 60 40 20 0 access metro/core Channels with viewers Channels broadcasted 10 100 1000 10000 100000 Waste of bandwidth, waste of energy! Selective pre-fetching of TV channels Instead of each node pre-fetching all TV channels, pre-fetch only a selection of channels Pre-fetch active channels (channels for which there are viewers) + a small number of inactive channels Room size = number of inactive channels prefetched Number of users Trace-driven simulation Results bandwidth savings Using huge dataset from a nationwide IPTV provider: 255k users, 6 months, 150 TV channels, 622 DSLAMs, 11 regions Number of channels prejoined 160 140 120 100 80 60 40 20 50% bandwidth reduction 33% bandwidth reduction 0 r = 0 r = 5 r = 10 r = 15 r = 20 r = 25 r = 0 (metro) All channels Results requests affected Is it worth it?, part 1 16% % of requests 14% 12% 10% 8% 6% 4% < 2% requests affected < 0.1% requests affected 2% 0% r = 0 r = 5 r = 10 r = 15 r = 20 r = 25 r = 0 (metro) All channels
Is it worth it?, part 1 Conclusion Today probably not, in the future probably yes Router power consumption model What about energy savings? Max power Power consumption l p Load Max capacity We assume the router can turn off ports not in use l and p based on real measurements [Chabarek et al.] 250 edge routers + 50 core routers Is it worth it?, part 2 Energy savings per year Millions 3.0 2.5 2.0 1.5 1.0 0.5 0.0 cost savings ( ) energy savings(kwh) today 2-4 years 10-15 years Conclusion Today not, in the future maybe 500 450 400 350 300 250 200 150 100 50 0 Thousands Cost savings per year
Zapping delay In IPTV this delay can add up to two seconds or more should be below 430ms[Kooij et al.] Several solutions proposed Video coding and processing techniques Network level Problems of existing solutions: Complexity Additional video servers needed Most zapping is linear... IPTV today IPTV NETWORK Requesting channel 3 Requesting channel 3 Sending channel 3 IPTV with channel smurfing IPTV NETWORK Channel smurfing Besides the channel requested, send N neighbouring channels concurrently for C seconds. Requesting channel 3 Requesting channel 2,3,4 Sending channel 2,3,4
Trace-driven simulation Results requests with no delay 100% Using huge dataset from a nationwide IPTV provider: 255k users, 6 months, 150 TV channels, 622 DSLAMs, 11 regions % of switching requests with no delay 90% 80% 70% 60% 50% 40% 30% 20% 10% 2 neighbours 4 neighbours 6 neighbours 8 neighbours 10 neighbours ideal predictor 0% 10 seconds 30 seconds 1 minute 2 minutes Always Concurrent channel time Results requests with no delay (zapping periods only) 40 Results average bandwidth % of switching requests with no delay 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 10 seconds 30 seconds 1 minute Concurrent channel time 2 neighbours 4 neighbours 6 neighbours 8 neighbours 10 neighbours ideal predictor Average bandwidth consumption (Mbps) 35 30 25 20 15 10 5 0 10 seconds 30 seconds 1 minute 2 minutes Always Concurrent channel time 2 neighbours 4 neighbours 6 neighbours 8 neighbours 10 neighbours ideal predictor Is it worth it? Very simple to implement Small software upgrade No additional video servers needed Performance close to optimal predictor Distance to ideal predictor 45% 40% 35% 30% 25% 20% 15% 10% 5% 0% 2 neighbours 4 neighbours 6 neighbours 8 neighbours 10 neighbours 10 seconds 30 seconds 1 minute 2 minutes Always Concorrent channel time
Take-home message 1. Computer science knowledge can be used in novel, practical, useful ways 2. It is important to build realistic scenarios to evaluate our ideas 3. It is fundamental to accept that any technical solution has limitations and that these should not be concealed 4. Be aware that a solution to a problem is only relevant if the benefits clearly outweigh the disadvantages Interested in these matters? Any idea what the G in GREEN could stand for? Feel free to contact me: fernando.ramos@cl.cam.ac.uk References [Cha et al.] M. Cha, P. Rodriguez, J. Crowcroft, S. Moon, and X. Amatriain. Watching television over an IP network. In Proc. ACM IMC, 2008. [Chabarek et al.] J. Chabarek, J. Sommers, P. Barford, C. Estan, D. Tsiang, and S. Wright. Power awareness in network design and routing. In Proc. IEEE INFOCOM, 2008. [Cisco] Cisco visual networking index: Forecast and methodology 2008-2013, 2008. [Kooij et al.] R. Kooij, K. Ahmed, K. Brunnstr om, and K. Acreo. Perceived quality of channel zapping. In proceedings of the IASTED, 2006. [Piper] B. Piper. United states IPTV market sizing: 2009-2013. Technical report, Strategy Analytics, 2009. [Qiu et al.] T. Qiu, Z. Ge, S. Lee, J. Wang, Q. Zhao, and J. Xu. Modeling channel popularity dynamics in a large IPTV system. In Proc. ACM SIGMETRICS, 2009. [Ramos et al. 1] F.M.V. Ramos, R.J. Gibbens, F. Song, P. Rodriguez, J. Crowcroft, I.H. White. Reducing energy consumption in IPTV networks by selective pre-joining of channels. Submitted to ACM SIGCOMM Workshop on Green Networking. F.M.V. Ramos, J. Crowcroft, R.J. Gibbens, P. Rodriguez, I.H. White. Channel Smurfing: Minimising Channel Switching Delay in IPTV Distribution Networks. IWITMA 2010 (To appear)